Skip to contents

Compute mvSuSiE-RSS TWAS weights from summary statistics

Source: R/regularized_regression.R
mvsusie_rss_weights.Rd

Multi-context summary-statistics analog of mvsusie_weights: extracts coefficients from an existing mvsusieR::mvsusie_rss fit, or fits one from stat$z (variants x conditions) and LD.

Usage

mvsusie_rss_weights(
  stat,
  LD,
  mvsusie_rss_fit = NULL,
  prior_variance = NULL,
  residual_variance = NULL,
  L = 30,
  L_greedy = 5,
  retain_fit = FALSE,
  ...
)

Arguments

stat

A list with z (matrix variants x conditions) and n (numeric vector or scalar).

LD

LD correlation matrix.

mvsusie_rss_fit

Optional pre-fitted mvsusie_rss object.

prior_variance

Optional mvSuSiE prior variance specification. When NULL, mvsusieR::create_mixture_prior() is used with R = ncol(stat$z).

residual_variance

Optional residual covariance matrix.

L

Maximum number of single effects (default 30).

L_greedy

Initial greedy effect count (default 5).

retain_fit

If TRUE, attaches the fitted object as an attribute.

...

Additional arguments forwarded to mvsusieR::mvsusie_rss.

Value

A numeric matrix of per-variant per-context weights (variants x conditions).

Details

Follows the *_rss_weights(stat, LD, ...) contract. Expects stat$z to be a numeric matrix (variants x conditions) and stat$n a per-context vector or scalar.